Rangi42
39e9315e8a
Use a smaller size for the lexer buffer
...
Large sizes are more efficient when it's actually buffered,
but most of the time `mmap` is used instead, and the extra size
just slows down allocation of lexer states.
2024-05-17 20:59:42 -04:00
Sylvie
e2633d5b66
Use the standard stream buffer size for the lexer buffer ( #1396 )
2024-05-17 11:02:36 +02:00
Rangi42
b8387427a6
Use consistent RGBDS_<PROG>_<NAME>_HPP header guard convention
2024-04-14 21:50:05 -04:00
Rangi42
106e516962
Split the .peek() method into its next-char and lookahead cases
2024-03-28 15:37:25 -04:00
Rangi42
6f74e4fb9c
Remove the suboptimal .canPeek() and .peek() methods
2024-03-28 14:49:36 -04:00
Rangi42
83c0634f15
Refactor peekInternal to be a LexerState method
2024-03-28 13:21:28 -04:00
Rangi42
4172d330b9
Refactor BufferedContent and Expansion to have methods
...
Use a buffer size that is a power of two for fast modulus
2024-03-28 12:57:23 -04:00
Rangi42
b6039870e5
Remove now-redundant MmappedContent struct
2024-03-27 11:27:34 -04:00
Rangi42
78801e324c
Group pointer and size as a ContentSpan struct
2024-03-27 11:27:34 -04:00
Rangi42
bf0cabb3ea
Use std::shared_ptr for lexer capture buffers
2024-03-27 11:27:34 -04:00
Rangi42
32db0a0f18
Rename CaptureBody to Capture, and refactor its methods
2024-03-26 12:29:59 -04:00
Rangi42
a167d23d01
Use content-specific destructors for lexer state
...
Also rename `LexerState` content structs from `*LexerState` to `*Content`
2024-03-26 12:29:59 -04:00
Rangi42
2b44672873
Rename a few variables
2024-03-25 14:22:38 -04:00
Rangi42
d9e5e57e27
Make CaptureBufs be parser values instead of a single static global
2024-03-23 19:44:54 -04:00
Rangi42
e9e8915725
Refactor to keep lexerState and lexerStateEOL static
...
Also run `clang-format` on everything
2024-03-23 19:14:46 -04:00
Rangi42
6a5518e0c5
Use RAII to unmap or close the lexer states' files automatically
2024-03-22 14:25:36 -04:00
ISSOtm
e255af9e10
Do not limit strings to 255 characters
2024-03-22 14:25:36 -04:00
ISSOtm
9f239f6dcc
Use std::shared_ptr<std::string> for lexed/parsed strings
2024-03-22 14:25:36 -04:00
Sylvie
0300971a17
Fix some header #includes with clangd LSP ( #1370 )
...
Co-authored-by: ISSOtm <eldredhabert0@gmail.com >
2024-03-21 15:13:10 -04:00
Rangi42
40db9d5cef
Use std::string for lexer state paths
2024-03-18 14:42:05 -04:00
Sylvie
53537cf9af
Remove now-unnecessary enum keyword ( #1338 )
...
C++ does not need it
2024-03-08 19:40:41 -05:00
Rangi42
097b7c3baf
Use std::string for symbol names
2024-03-07 12:40:33 -05:00
Rangi42
bf45ebb178
Run clang-format to fix some inconsistent style
2024-03-07 11:52:53 -05:00
Sylvie
292adb27a3
Adapt the RGBASM parser to C++ ( #1333 )
...
This uses variants instead of a `%union`, and "complete symbols"
that can call complex constructors.
2024-03-06 20:40:36 -05:00
Sylvie
e74073e480
Run clang-format on everything ( #1332 )
2024-03-04 14:22:49 -05:00
Sylvie
8cf446b14c
Use std::variant for lexer mmap/buffer state ( #1328 )
2024-03-03 18:45:50 -05:00
Rangi42
277ea9be28
Replace RGBASM non-null pointers with references
2024-03-03 00:57:03 -05:00
Rangi42
19bb12754b
Use new allocation for expanding \#
2024-03-02 08:04:59 -05:00
Rangi42
dbcb82799e
Use std::vector for capture buffer
2024-03-02 06:46:11 -05:00
Sylvie
a71e4086a2
Use std::string_view for macro bodies ( #1326 )
...
This removes the last use of `strdup`
This required making a lot of related pointers be `const`.
That in turn conflicted with the need to `munmap()` a pointer
eventually, which was similar to the need to eventually `free()`
an `Expansion`'s contents, so I used the same solution of a
`union`. That lets us normally use the `const` pointer for
`const` correctness, and the non-`const` one for the not-really-
mutating destruction cases.
2024-03-02 13:21:28 -05:00
Rangi42
b488d3a90f
Use std::optional<std::string> for lexer expansion names
2024-03-02 04:34:42 -05:00
Rangi42
e14ba664ea
Remove redundant (void) parameter declarations
2024-03-01 10:41:47 -05:00
Rangi42
91d22f180e
Reorganize the asm parser to be more like the linker script parser
...
Split the declarations into those required for the `%union` and
those only required for the `code`.
Only declare functions on top; define them at the bottom.
2024-02-29 16:23:53 -05:00
Sylvie
043db49676
Replace NULL with nullptr ( #1321 )
2024-02-29 15:06:33 -05:00
Sylvie
eff8c324c8
Remove now-unnecessary struct keyword ( #1320 )
...
C++ acts like structs are `typedef`ed by default
We do have to keep `struct stat`, since there's ambiguity
with the function also called `stat`.
2024-02-29 14:41:58 -05:00
Rangi42
d1fa5ccd4d
Rename lexer_DeleteState to lexer_CleanupState
2024-02-29 13:44:08 -05:00
Rangi42
af055ecd27
Use automatic allocation for IF stacks and expansions
...
Switch to using `std::deque` for IF stacks (supports `.clear()`)
2024-02-29 13:44:08 -05:00
Rangi42
31836967fa
Use automatic allocation for lexer states
...
Lexer states are now owned by fstack contexts
2024-02-29 13:44:08 -05:00
Rangi42
843f3394c8
Use automatic allocation for DS args
2024-02-29 13:44:08 -05:00
Rangi42
b1aa98b43d
Use automatic allocation for PURGE args
2024-02-29 13:44:08 -05:00
Sylvie
595c066c2a
Remove #include <stdbool.h> ( #1317 )
2024-02-24 10:51:46 -05:00
Sylvie
54d6a22d19
Build with pedantically standard C++ ( #1309 )
...
* Remove array designators (not standard C++)
* Build with pedantically standard C++
2024-02-23 16:46:53 -05:00
Rangi
756f2866bb
Refactor alignment spec parsing ( #1253 )
2023-11-25 00:06:05 +01:00
Rangi
46e29de66f
Implement ds align[alignment, offset] ( #1181 )
2023-11-21 23:57:47 +01:00
Rangi
1e70e703a7
Build everything as C++ ( #1176 )
2023-11-07 21:45:56 +01:00
